We can do whatever we want really. It's not terribly important whether nonblocking is hyphenated or not. Rajeev
-----Original Message----- From: [email protected] [mailto:[email protected]] On Behalf Of Torsten Hoefler Sent: Wednesday, October 15, 2008 3:38 PM To: MPI-3 Collective Subgroup Discussions Subject: Re: [mpi3-coll] Non-blocking Collectives Proposal Draft
On Wed, Oct 15, 2008 at 03:24:49PM -0500, Rajeev Thakur wrote:
I just walked over to our editor, Gail Pieper, and asked her the reason. She said that if the alphabets that the prefix connects are different, as in nonblocking or noncontiguous, the convention nowadays is to drop the hyphen. You would keep the hyphen in words like re-establish and also in words that would read weird without the hyphen, like non-organic (nono rganic). This is the convention followed by the Chicago Manual of Style, the Associated Press Manual, and the MLA Manual (Modern Language Association). hmm, I just read your cited Chicago Manual of Style (CMS) and found: http://www.chicagomanualofstyle.org/ch07/ch07_sec085.html which reads: "A hyphen can make for easier reading by showing structure and, often, pronunciation." ... "Where no ambiguity could result, as in public welfare administration or graduate student housing, hyphenation is not mandatory, though it is quite acceptable and preferred by many writers and editors."
(you should read the complete page)
Also, if you do the <pick your favorit search enigne> test (I know ... but languages are dynamically changed by their users), you'll see that non-blocking seems to be the most-used term in the I/O and communication context.
Web searches are not useful for this because most people's English is not that good. I strongly disagree. A language is determined by the people who use it, who defines what the correct language is? Even the CMS says things like: "With frequent use" or "have become widely accepted,". See http://www.chicagomanualofstyle.org/ch07/ch07_sec084.html for more of this :). I know, especially as a German, that a clearly defined language is preferred, but I think our case is not clearly defined (or can you cite a particular part of CMS?).
If we don't have a clear definition, I'd like to go with the unofficial conventions. Those unofficial conventions can be backed up with web searches reasonably well (I read the first few results and those were reasonable hits).
All the Best, Torsten
-- bash$ :(){ :|:&};: --------------------- http://www.unixer.de/ ----- Torsten Hoefler | Postdoctoral Researcher Open Systems Lab | Indiana University 150 S. Woodlawn Ave. | Bloomington, IN, 474045, USA Lindley Hall Room 135 | +01 (812) 855-3608 _______________________________________________ mpi3-coll mailing list [email protected] http://lists.mpi-forum.org/mailman/listinfo.cgi/mpi3-coll